Download Summary

Summary
In this thesis, we explored confounders of measurements by the transpulmonary
thermodilution technique, in patients with intrathoracic pathology (chapter 3) or valvular
insufficiencies (chapter 5). The possible role of mathematical coupling explaining the
superiority of volumes over pressures as indicator of preload and the influence of cardiac
function on preload parameters is described in chapter 4 and 6, respectively. Furthermore,
we evaluated whether two non-calibrated pulse contour cardiac output techniques track
thermodilution cardiac output changes during altering preload and afterload, in patients after
cardiac surgery (chapter 8 and 9).
Chapter 1
In the critically ill, hemodynamic monitoring is usually performed with a pulmonary artery
catheter (PAC), allowing assessment of central venous pressure (CVP) and pulmonary
artery occlusion pressure (PAOP), both which are assumed to be reliable guides for fluid
therapy. Moreover, cardiac output can be measured via thermodilution. However, CVP and
PAOP measurements are influenced by lung compliance and intrathoracic pressures.
Moreover, the PAC is invasive with inherent risks. To bypass these flaws, a less invasive
technique was developed which is able to measure volumes instead of pressures, not
influenced by positive pressure ventilation. However, there are some drawbacks of the
transpulmonary thermodilution method, inherent to technique. Indeed, in patients with
valvular insufficiencies or intrathoracic pathology, measurements can be confounded.
Moreover, the contribution of mathematical coupling to the superiority of volumes over
pressures as indicators of preload is unclear, when both cardiac output and cardiac volumes
are derived from the same thermodilution curve. These doubts form the basis of the first aim
of this thesis: to clarify some issues associated with the transpulmonary thermodilution
technique in assessing cardiac output, preload and fluid responsiveness. In the critically ill
both preload and afterload conditions of the heart change over time, due to altered cardiac
compliance, volume status, change of inotropic or vasodilator doses and ventilator settings,
which will result in CO changes. In the second part of this thesis we evaluate whether two
minimally invasive, non-calibrated arterial waveform based techniques, the modified
ModelFlow and FloTrac/Vigileo, track these CO changes.
Chapter 2
We give a summary of the literature of the role intrathoracic blood volumes determined
by transpulmonary thermodilution, in the monitoring of critically ill patients, with respect
to assessing volume status, heart function and response to fluids in the treatment
of hypovolemia and shock. We concluded that functional hemodynamic monitoring of
intrathoracic volumes with double or single indicator dilution is a useful tool to assess
preload and fluid responsiveness in critically ill patients. However, the effect of volumetric
(rather than pressure) monitoring on morbidity and mortality of critically ill patients
is lacking, although the method is less invasive compared to the PAC.
Chapter 3
We describe two patients, one with severe haemorrhage and one with a partial anomalous
pulmonary vein in which cardiac output measurements were performed simultaneously
by means of the single-indicator transpulmonary thermodilution technique and continuous
pulmonary artery thermodilution method. In both cases, the methods revealed clinically
significant different cardiac output values based upon the site of measurement and the
underlying pathology. In the first patient with excessive blood loss, cold indicator could be
lost due to the accumulation of blood in the thorax. Vliers et al.1 demonstrated the leave and
re-entrance of indicator from the vascular bed into the lungs by recording the intrabronchial
air temperature changes during the passage of the cold indicator through the pulmonary
vascular bed. This will lead to a reduction and distorted thermodilution curve, resulting in
overestimation of the cardiac output. However, in the second patient a persistent anomaly
was present. In this patient, both techniques, for cardiac output measurement, seem to
measure the real-time cardiac output of, respectively, the left and right ventricle, since the
amount of blood passing the thermistor of the pulmonary artery catheter is the sum of the
shunt flow and of the vena cava inferior and superior flow.
Chapter 4
This study investigated the role of mathematical coupling of shared measurement error
when cardiac volumes and cardiac output are obtained from the same thermodilution
curve. Eleven consecutive and mechanical ventilated patients with hypovolemie after
coronary artery surgery and a pulmonary artery catheter in place, allowing continuous
cardiac index (CCIp) measurements, received a femoral artery catheter for transpulmonary
thermodilution measurements (PiCCOTM, CItp). A total 48 fluid loading steps of 250 mL
were done. We found that central venous pressure and volumes equally related to CItp and
CCIp and that fluid responses were predicted and monitored similarly by CItp and CCIp.
However, during fluid loading, changes in volumes related better to changes in CItp when
derived from the same curve than to changes in CItp of an unrelated curve or changes in
CCIp. These data suggest that the superiority of volumes as indicator of fluid responsiveness
over filling pressures can be caused, at least in part, by mathematical coupling of shared
measurement error when cardiac volumes and output are derived from the same
thermodilution curve.
Chapter 5
This study investigated whether residual left-sided valvular insufficiencies after valvular
surgery confound transpulmonary thermodilution cardiac output (COtp) technique. We
compared the technique with the continuous right-sided thermodilution technique (CCO) after
valvular and coronary artery surgery. After valvular surgery, there was minimal aortic
insufficiency in 4 patients and minimal to moderate mitral valve insufficiency in 6. 5 Fluid
loading steps (250 mL) were done in each patient. We found a lower cardiac output after
valvular than coronary artery surgery but responses to fluid loading steps were similar among
surgery types and techniques. We also found that after valvular and coronary artery surgery,
cardiac output was lower prior to responses than in nonresponses to fluids, by either
technique. After valvular surgery, COtp and CCO correlated but fluid-induced changes did
not. After coronary artery surgery, COtp and CCO correlated and changes also did. At fluidinduced CCO increases <20%, the r for changes in cardiac output measured by both
techniques was similar after valvular and coronary artery surgery. These data imply that
COtp and CCO are of similar value in predicting and monitoring fluid responses after both
surgery types, therefore arguing against left-sided valvular insufficiencies, severely
confounding COtp.
Chapter 6
In this chapter we compared the value of cardiac filling pressures and volumes in patients
after valvular (VS) and coronary artery surgery (CAS) patients, since cardiac function may
differ after both types of surgery and therefore may effect assessment of fluid
responsiveness. Eight patients after VS and 8 after CAS were included. In each patient, five
sequential fluid loading steps of 250 mL of colloid were done. Fluid responsiveness was
defined by a cardiac index (CI) increase >5% or ≥10% per step. We found a lower global
ejection fraction and a higher pulmonary artery pressure (PAOP) after VS compared to CAS.
In responding steps after VS, PAOP and volumes increased, while central venous pressure
(CVP) and volumes increased in responding steps after CAS. After VS, baseline PAOP as
well as changes in PAOP and volumes were of predictive value and PAOP and volume
changes equally correlated to CI changes. After CAS, CVP and volumes were of predictive
value and changes in CVP and volumes correlated to those in CI. These data suggests an
effect of systolic cardiac function on optimal parameters of fluid responsiveness and
superiority of the pulmonary artery catheter over transpulmonary dilution for hemodynamic
monitoring of VS patients.
Chapter 7
In this chapter we summarize the physiology of static indicators used to predict and
monitor preload responsiveness in critically ill patients on mechanical ventilation. First,
we explain the definitions preload and fluid responsiveness used in the literature. Secondly,
we discuss filling pressures and volumes of the heart and the influence of positive
pressure mechanical ventilation on both measurements. In the last part, we describe
the influence of contractility and afterload of the heart on the cardiac function curve
relating cardiac output to end-diastolic filling pressure. We also explain the relative
predictive value of volumes versus pressures for changes in cardiac output, depending
on the position on the compliance curve and the position and shape of this curve.
Furthermore, we illustrate the clinical implications of these physiological considerations.
We concluded that the relative predictive value of static filling pressures and volumes
for the cardiac output response to preload changes depends on biventricular systolic
and diastolic function. Overall, this implies that both pressures and volumes may be
needed to predict and monitor fluid responses. However, the ultimate guide to fluid
therapy may be frequently intermittent or continuous cardiac output monitoring to prevent
fluid overloading by discontinuing fluid infusion when cardiac output does not further
increase.
Chapter 8
This chapter reports on our prospective study to investigate whether changes in less
invasive, non-calibrated pulse-contour cardiac output (by modified ModelFlow, COmf )
and derived stroke volume variation (SVV), as well as systolic and pulse pressure
variations predict changes in bolus thermodilution cardiac output (COtd), induced
by continuous and cyclic increases in intrathoracic pressures by increases in positive
end-expiratory pressure (PEEP) and tidal volume (Vt), respectively. We found that SVV
increased at PEEP 10cmH2O and 15 cmH2O, concomitantly with a decrease in COmf
and COtd. In contrast, systolic and pulse pressure variation did not increase. We also found
that changes in COmf correlated with those in COtd, whereas changes in SVV did not.
Remarkably, variables did not change when Vt was increased up to 50%. Thus, in
mechanically ventilated patients after cardiac surgery, during continuous increases in
intrathoracic pressure, tracking a fall in COtd by COmf is more sensitive than a rise in
SVV, which is more sensitive than systolic pressure variation and pulse pressure variation.
These result suggest that PEEP causes a reduction in biventricular preload and is the main
factor in decreasing cardiac output and increasing SVV.
Chapter 9
In this prospective study we investigated whether cardiac output and its postoperative
course by a new non-invasive pulse-contour technique of radial artery pressure waves,
without need for calibration (FloTracTM/VigileoTM [FV], Edwards Lifesciences) conforms
with the standard bolus thermodilution method via a pulmonary artery catheter. Fiftysix
simultaneous measurement sets were obtained, in 20 patients up to 24 h after cardiac
surgery. We found that cardiac output measured by the non-invasive pulse-contour FV
method reasonably well compares to that measured by bolus thermodilution via pulmonary
artery catheter, in cardiac surgery patients, over a relatively wide range of cardiac output.
Moreover, for clinical purposes the concordance of detected changes in cardiac output over
time was satisfactory. Thus, the FV pulse-contour method is a clinically applicable noninvasive pulse-contour method for cardiac output assessment without the need for
calibration, after cardiac surgery.